Hey just want to preface this by stating that I am new to this type of online platform of experiment building. Typically, we have our experiments in the lab run in house, but due to restrictions in place it is not possible so we are branching into new avenues.
My question is that I am having trouble importing previous experiments that were made with psychopy builder, or were just python coded and imported into the psychopy coder portion of it. My builder experiments are having issues with finding the images.
name : ‘image_4’, units : ‘pix’,
image : ‘Slide5.jpg’,
when initializing my images for instructions and other objects. I have the image files in the same project folder on gitlab but it does not seem to find the images. Am I missing a step in terms of forking over the files or do I need to dictate a specific file path due to the online nature of it?
As for my coder built experiments. I tried dropping them in to psychopy and used the “sync with web project” option and got a “403 Forbidden. nginx” error when trying to pilot the experiment. Not sure what I did wrong for that one. Any help at all for either or both of these situations is greatly appreciated!
Unfortunately you cannot present python coded scripts online. Experiments will need to be made in psychopy builderview to get online (since this view has the tools to also write your experiment to java script, which is needed to run in the browser)
For your builder experiments, it is easiest if you can share a screenshot of the error you encounter when running online and, if possible, share the URL to the gitlab project with us (from your experiment page on pavlovia go >view code> settings> permissions> set visibility to public THEN share the URL for the gitlab project with us). Is your issue a missing resource error? if so have you tried adding resources manually? e.g. Shuffle order of multiple routines
Previously, I would be running the experiment with the needed picture files in the same folder and would just need to tell psychopy the file name, and it would search in its immediate vicinity for the object that fits its name. But what I understand that you are saying is that in the experiment builder I need to delegate a different path to access the jpgs?
As for the coding, only javascript coding will work? Does that include if I were to add a piece of coding via the “custom -> code” component on the psychopy builder?
After messing around for a bit, I managed to fix my image problem using the method in the link you provided which is much appreciated!
The issue now is my data files, which do not seem to have the same layout of my previous CSV files, is there anyway to ensure the data being saved is in the format similar to previous? The builder data file. _untitled_2020_Nov_02_1349.csv (7.7 KB)
The online experiment file test_expA_2020-11-02_13h57.42.882.csv (3.3 KB) .
Yes that is the correct URL to share - pleased you fixed you original issue. For the data file format, the formatting will be slightly different for online experiments (there are multiple variables more relevant to online exps e.g. operating system etc) but it looks as though the same data is available in both of these files, just in different columns?